JOURNAL OF VIROLOGY, Jan. 1989, p. 392-397 Vol. 63, No. 10022-538X/89/010392-06$02.00/0Copyright © 1989, American Society for Microbiology

Abortive Replication of Choleraphage F149 in Vibrio choleraeBiotype El Tor

RUKHSANA CHOWDHURY, SIDDHARTHA K. BISWAS, AND JYOTIRMOY DAS*

Biophysics Div,ision, Indian Institute of Chemical Biology, 4 Raja S.C. Mullick Road, Calcutta 700 032, India

Received 14 April 1988/Accepted 17 September 1988

Choleraphage 4)149 adsorbed irreversibly to Vibrio cholerae biotype el tor cells, and 50% of the injectedphage DNA bound to the cell membrane. Although no infectious centers were produced at any time duringinfection, the host macromolecular syntheses were shut off and the host DNA underwent chloramphenicol-inhibitable degradation. Synthesis of monomeric phage DNA continued similar to that observed in thepermissive host. However, the concatemeric DNA intermediates produced were unstable and could not bechased to mature phage DNA. Pulse-labeling of UV-irradiated infected cells at different times during infectionallowed identification of phage-specific proteins made in this nonpermissive host. Although most of the earlyproteins were made, only some of the late proteins were transiently synthesized.

Vibrio cholerae biotype classical and Vibrio choleraebiotype el tor are biotypes of the causative agent of thediarrheal disease Asiatic cholera. These biotypes are iso-genic and are not easily distinguishable by biochemical tests(8). Both V. cholerae and V. cholerae biotype el tor reactwith anti-V. cholerae antisera, and taxonomic studies haveshown that these two biotypes are the same species (4, 5).One of the differentiating criteria between these cells is theirsusceptibility to the group IV choleraphages. While all theother groups of choleraphages can infect and lyse all strainsof the el tor biotype, phages belonging to group IV can infectand lyse all strains of the classical biotype but none of the eltor strains (12). The mechanism of inhibition of group IVphage infection in V. cholerae biotype el tor cells is notknown, and spontaneous phage receptor mutants of classicalvibrios have often been mistaken for V. cholerae biotype eltor cells. Most of the el tor strains isolated and characterizedso far carry a phage called kappa (18); initially it wasassumed that the presence of phage kappa interferes in someunknown way with the replication of group IV phage in V.cholerae biotype el tor cells. However, this assumption isquestionable, since nonlysogenic V. cholerae biotype el torcells also fail to support the growth of group IV cholera-phages (10, 12).The process of infection of V. cholerae biotype classical

cells by phage (149, a representative strain of group IVcholeraphages, has recently been examined in detail. Phage(149 is a large virus (1) and contains a linear, double-stranded DNA molecule of 69 megadaltons, equivalent to102 kilobase pairs. The DNA molecules are a limited set ofcircular permutation of the phage genome and have severalsingle-strand interruptions along their length that are repair-able by DNA ligase (17). The phage DNA codes for 26 earlyproteins and 23 late proteins (15). The intracellular replica-tion of 4149 DNA involves a concatemeric DNA structureserving as the substrate for the synthesis of mature phageDNA, which is eventually packaged by a headful mechanismstarting from a unique pac site in the concatemeric DNA (3,10). The growth of this phage is extremely sensitive to theconcentration of phosphate ions, and no phage growthoccurs in medium containing more than 0.1% phosphate(15). After infection under high-phosphate conditions, the

* Corresponding author.

concatemeric DNA intermediates are not formed, althoughsynthesis of monomeric molecules is unaffected (3).Choleraphage (138, belonging to group II and containing

a linear double-stranded circularly permuted DNA molecule,can replicate in V. cholerae biotype el tor cells (2, 12). It hasbeen reported that, as in phage F149, a concatemeric DNAintermediate is involved in the replication of phage (138 (2).Thus, formation and stabilization of concatemeric DNAstructures essential for the synthesis of mature phage DNAand their packaging into phage heads are not hindered in V.cholerae biotype el tor cells for the replication of phage(138. Incidentally, the replication of this phage is insensitiveto the concentration of phosphate ions in the growth me-dium, in contrast to what has been observed for phage (149(2, 15). It thus appears that a certain host gene functionmight be required for the replication of phage (1149 and notof phage (1138 and that this function is lacking in V. choleraebiotype el tor cells.

In this context, in the present report we describe experi-ments that examine the process of infection of V. choleraebiotype el tor cells by phage (149, with a view to determin-ing how far phage replication can proceed in this nonpermis-sive host. The results presented here show that after infec-tion of V. cholerae biotype el tor cells, phage (149 DNAmakes an abortive attempt to form the concatemeric DNAintermediate. However, these structures cannot be stabi-lized in these cells and, as such, cannot produce infectiouscenters.

MATERIALS AND METHODSBacterial and phage strains. The V. cholerae and cholera-

phage strains used in this study are listed in Table 1. Allbacterial and bacteriophage strains were obtained from theCholera Research Centre, Calcutta, India. V. choleraeOgawa 154, the universal host for the propagation of cholera-phages, was used for phage propagation; V. cholerae 569Band V. cholerae biotype el tor strain MAK757 were used asthe hosts for phage infection studies. Cultures were main-tained as described previously (9, 16). Cell and phage growthwere assayed on nutrient agar plates. High-titered phagestocks and 32P-labeled phages were prepared by infecting V.cholerae 154 cells with phages as described previously (15).Media and buffers. The media used for growing bacteria

were prepared as described previously (7). The Tris-Casa-

392

ABORTIVE 4P149 REPLICATION IN V. CHOLERAE BIOTYPE EL TOR

TABLE 1. Bacterial strains and choleraphages

Bacteria andBacteriophagesdDescription Referencebacteriophages

V. cholerae569B Prototroph, Tox+, classical biotype, 8

Inaba serotype154 Prototroph, classical biotype, Ogawa 11

serotypeMAK757 Prototroph, el tor biotype, Ogawa 11

serotype

Bacteriophages(P149 Serological group IV, linear 17

double-stranded, circularlypermuted DNA of 69 megadaltons(102 kilobase pairs)

(D138 Serological group II, linear 2double-stranded, circularlypermuted DNA of 30 megadaltons(45 kilobase pairs)

mino Acids-glucose medium containing 0.4% KH2PO4 usedfor cell growth is referred to herein as HP medium, whereasthe same medium used for phage growth but containing0.04% KH2PO4 is referred to as LP medium. Label-termi-nating buffer contained 50 mM Tris hydrochloride (pH 7.5),5 mM MgCl2, 2 mM NaN3, and 3 mg of thymidine per ml.

Isolation of phage DNA. The phage DNA was isolated frompurified phages as described previously (3).

Preparation of infected cell lysates. Cells in the logarithmicphase of growth (2 x 108 to 3 x 108 CFU/ml) in HP mediumwere infected with phage (P149 at a multiplicity of infection(MOI) of 10. Two minutes was allowed for adsorption, andthen the infected cells were washed and suspended in LPmedium. At different times during infection, samples of theinfected culture were withdrawn and, when required, labeledwith 20 puCi of [3H]thymidine (specific activity, 18.8 Ci/mmol; Bhabha Atomic Research Centre, Trombay, India)per ml for the desired length of time. Labeling was termi-nated by adding an equal volume of label-terminating buffer,and the sample was sedimented (10,000 x g, 10 min, 4°C)and then lysed by suspension in solution (1/20 the culturevolume) containing 0.01 M Tris hydrochloride (pH 7.9),0.001 M EDTA, and either 1% sodium dodecyl sulfate or 2%Sarkosyl NL97. The suspension was incubated at 37°C for 30min before being layered on neutral 5 to 20% (wt/vol)sucrose gradients.

Sucrose gradient centrifugation. Infected cell lysates were

analyzed by velocity sedimentation in neutral 5 to 20% (wt/vol) sucrose gradients in a Sorvall AH650 rotor at 30,000 rpmfor 150 min at 15°C. Fractions were collected, precipitatedwith trichloroacetic acid, filtered, washed, dried, and as-

sayed for radioactivity as described previously (2).Analysis of fast-sedimentation complex. Cells infected with

32P-labeled (P149 (specific activity, 2.5 x 105 cpm /PFU) atan MOI of 10 in LP medium were centrifuged (10,000 x g, 5min) after 2 min at 4°C, washed with cold medium, andsuspended in fresh prewarmed medium. At different times,samples were withdrawn, labeled with 25 ,uCi of [3H]thymi-dine per ml for the desired length of time, and analyzed byvelocity sedimentation in sucrose gradients formed over a

shelf of 1.3 g of CsCl per ml in 40% sucrose as describedpreviously (3).

Phage-specific protein synthesis. Host cells in the logarith-mic phase of growth were harvested by centrifugation(10,000 x g, 5 min), suspended in dilution buffer, and

irradiated with UV light at a dose rate of 1 J/m2 per s andused for infection with phage 4149 as described previously(15). At different times during infection infected cells werepulse-labeled with 10 ,uCi of [35S]methionine (specific activ-ity, >800 Ci/mmol; Amersham International) per ml, and thenewly synthesized proteins were analyzed by sodium dode-cyl sulfate-polyacrylamide gel electrophoresis and autoradi-ography as described previously (15).

RESULTS

Preliminary considerations. V. cholerae biotype el torstrain MAK757, which does not carry the phage kappa, wasused as the host for infection by phage 4149. When phage0149 was added to these cells at the logarithmic phase ofgrowth at an MOI ranging from 1 to 10, more than 95% of theadsorbed phages were irreversibly bound to the cells. Al-though no infectious centers were detected in the cell pellet,cell growth was completely inhibited after infection, and notmore than 5% of MAK757 cells retained their colony-forming ability for infection at an MOI of 10.To investigate whether the adsorbed phages had injected

their DNA, cells were infected with 32P-labeled phages at anMOI of 10 and washed repeatedly to remove unadsorbedphages; acid-precipitable radioactivity was assayed in thecell pellet at different times during infection. More than 90%of the acid-precipitable radioactivity was recovered in thecell pellet until 60 min of infection. These results suggest thatthe injected 0149 DNA was not degraded in V. choleraebiotype el tor cells.

Fate of parental phage DNA. For infection of V. choleraebiotype classical cells by phage (149, the parental DNAbinds to several sites on the cell membrane (15). To examinewhether (149 DNA also binds to the cell membrane of V.cholerae biotype el tor cells, 32P-labeled phages were usedfor infection, and at different times infected cells were lysedby freezing and thawing. The lysates were analyzed bysedimentation through neutral sucrose gradients as de-scribed in Materials and Methods. About 50% of the input(P149 DNA became associated with the cell membraneimmediately after infection (Fig. la), and there was no lossof membrane-associated parental label at any time duringinfection, unlike that observed for infection of V. choleraebiotype classical (3). Furthermore, the total parental labelremained almost constant up to 60 min of infection. At latertimes during infection part of the freshly sedimenting phageDNA sedimented as heavier than the monomeric units (Fig.lb).Host DNA degradation. Since the cell viability was lost

after infection of V. cholerae biotype el tor with phage (149,macromolecule synthesis by the host was examined afterinfection. Cells of strain MAK757 were labeled with [3H]-thymidine, washed, starved in buffer for 30 min to exhaustthe cellular nucleotide pool, and infected with phage (149 atan MOI of 10. Within 15 min after infection more than 60%of host DNA was degraded (Fig. 2), as measured by the lossof trichloroacetic acid-precipitable radioactivity. The in-crease in acid-precipitable counts observed 20 min afterinfection presumably reflects synthesis of phage-specificDNA by utilizing the host DNA degradation products. It isknown that phage (149 utilized primarily the host DNAdegradation products for its own DNA synthesis (15). Wheninfected cells of strain MAK757 were treated with 200 ,ug ofchloramphenicol per ml at the time of infection not more

than 10% of the host DNA was degraded until 65 min afterinfection. The result suggests that some phage-specific pro-

VOL. 63, 1989 393

394 CHOWDHURY ET AL.

E

r4)

0T

0 10 20Fraction number

8

4Ea.0

C\JO No

4 -

2

0

FIG. 1. Distribution of (149 DNA in fast- and free-sedimentingcomponents at different times during infection. Cells in the logarith-mic phase of growth were infected with 32P-labeled phage (149 (0)at an MOI of 10, and the unadsorbed phages were removed byrepeated washing of the infected cell pellet. At different times afterinfection, samples were labeled with 25 p.Ci of [3H]thymidine per ml(0) for 5 min. The labeling was terminated, and the cells were

washed, lysed, and analyzed on a 5 to 20%o sucrose gradient formedover a CsCl shelf as described in the text 2 min (a) and 30 min (b)after infection. The sedimentation is from right to left.

teins responsible for host DNA degradation are synthesizedin infected cells of V. cholerae biotype el tor.

Pulse-labeled intracellular DNA after 1D149 infection. It hasbeen reported that replication of (149 DNA in the permis-sive host V. cholerae biotype classical involves (i) synthesisof monomeric molecules, (ii) conversion of monomeric mol-ecules to concatemeric DNA intermediates, and (iii) forma-tion of mature phage DNA by using concatemeric DNA as asubstrate (3). To examine the replication of (149 DNA in V.cholerae biotype el tor, cells were infected with phage (149at an MOI of 10 and labeled for 5 min with [3H]thymidine atdifferent times during infection. At the end of the labelingperiod cells were lysed, and the crude lysate was analyzedby velocity sedimentation in neutral sucrose gradients.Up to about 15 min of infection most of the pulse-labeled

DNA cosedimented with 32P-labeled (149 DNA used asmarker (Fig. 3). Phage DNAs synthesized 25 min afterinfection were of higher molecular weight as compared withthe monomeric units and were distributed in a wider peak(Fig. 3). This high-molecular-weight DNA intermediate ap-

parently contained molecules of widely varied sizes. Until 60min of infection, the synthesis of this high-molecular-weightDNA continued at a reduced rate (Fig. 3). (For infection ofV. cholerae cells by phage (149, phage DNAs synthesizedat 25 to 30 min after infection were predominantly in theconcatemeric form, with an average size of five to sixgenome equivalent lengths. These concatemers eventuallyled to the formation of monomeric units [3].) For infection of

0 5 10 15 20 25 30Time after infection (min)

FIG. 2. Degradation of V. cholerae biotype el tor cellular DNAafter phage (149 infection. Cells grown to the logarithmic phase inmedium containing 10 ,uCi of [3H]thymidine per ml and deoxyaden-osine (250 ,ug/ml) were washed, starved for 30 min in dilution bufferto deplete the pool of internal nucleotides, and then infected withphage 1449 at an MOI of 10 (0). A part of the sample was treatedwith 200 j±g of chloramphenicol per ml at the start of infection (A).At different times samples were removed, and acid-precipitableradioactivity was assayed. Uninfected, labeled cells were alsoexamined as a control (0).

el tor cells with phage (149, however, even 60 min afterinfection, none of the newly synthesized DNA moleculeswere resolved as monomeric units in the gradient (Fig. 3).Thus, the high-molecular-weight DNA intermediates formedduring infection of V. cholerae biotype el tor cells by phage(149 cannot serve as substrates for the synthesis of maturephage DNA. To investigate this possibility, infected cellswere pulse-labeled for 5 min with [3H]thymidine at a timeduring infection when high-molecular-weight DNA interme-

i12-

EQ.

08

0

0 5 10 5 20 25Fraction number

FIG. 3. Velocity sedimentation analysis of phage DNA interme-diates during intracellular replication. V. cholerae biotype el torMAK757 cells in the logarithmic phase of growth were infected with(149 at an MOI of 10 as described in the text. At 10 (0), 25 (0), and60 (A) min after infection, infected cells were labeled for 5 min with25 ,uCi of [3H]thymidine per ml, and the cell lysates were analyzedin 5 to 20% neutral sucrose gradients. The arrow indicates theposition of 32P-labeled (149 DNA in the gradient which was used asmarker. The sedimentation is from right to left.

J. VIROL.

ABORTIVE 0149 REPLICATION IN V. CHOLERAE BIOTYPE EL TOR

121

E8~~~~~~~~~~

CL

4-~~ ~ ~ ~~~~~

0 5 10 15 20 25 30

Fraction numberFIG. 4. Velocity sedimentation analysis of (149-infected V.

cholerae biotype el tor strain MAK757 cells pulse-labeled for 5 minwith 25 ,uCi of [3H]thymidine per ml at 25 min after infection. Thecell lysates were analyzed in 5 to 20% neutral sucrose gradients. Thepositions of monomeric and concatemeric (149 DNA in the gradientare indicated by the letters M and C, respectively. The sedimenta-tion is from right to left.

diate synthesis predominates (25 min after infection), and thepulse-labeled DNA was chased by suspending infected cellsin unlabeled medium.

Pulse-chase experiment. At the start of the chase, thepulse-labeled DNA sedimented faster than the monomericunits (Fig. 4). During the 40-min chase no conversion of thisDNA intermediate to monomers was observed. Instead, theacid-precipitable radioactivity in the high-molecular-weightDNA was reduced gradually. Thus, (149 DNA makes anabortive attempt to form the concatemeric DNA intermedi-ate essential for the synthesis of the monomeric units, whichare eventually packaged. These defective concatemers are

unstable and cannot serve as substrates for the synthesis ofmature phage DNA.

Site of progeny phage DNA synthesis. The analysis ofmembrane-associated and free-sedimenting components ofphage DNA of infected cells was repeated (Fig. 1), exceptthat at different times during infection [3H]thymidine was

added 5 min before harvesting. This enabled the determina-tion of distribution of parental 32P-labeled and nascent3H-labeled DNAs (Fig. 1).

Until 10 min after infection most of the newly synthesizedDNA sedimented freely at the position of the marker (149DNA (Fig. la). As with the classical vibrios, at 30 min afterinfection newly synthesized free-sedimenting DNA was

heavier than the mature phage DNA (Fig. lb). However, notmore than 10% of the newly synthesized DNA was recov-

ered in the fast-sedimenting complex. During infection in V.cholerae biotype classical cells, more than 50% of the newlysynthesized DNA was recovered as a fast-sedimenting mem-brane-associated complex (3). Thus, it appears that thedefective concatemeric DNA synthesized during infection inV. cholerae biotype el tor cells fails to associate with the cellmembrane. The free-sedimenting concatermeric DNAs were

eventually degraded. Membrane association of newly syn-thesized concatemeric DNA might be responsible not onlyfor its stability but also for subsequent cleavage into mono-mers and packaging into phage heads.

Synthesis of phage-specific proteins in V. cholerae biotype el

tor cells. In view of the fact that a phage-specific earlyprotein of molecular weight 64,500 is required for the forma-tion of the concatemeric DNA replicative intermediate (3,15), it was intriguing to examine the synthesis of phage-specific proteins in phage (149-infected V. cholerae biotype

el tor cells. The cells were UV irradiated to reduce cellularprotein synthesis, infected with phage (149, and labeled atdifferent times after infection with [35S]methionine for 10min. The newly synthesized proteins were analyzed bypolyacrylamide gel electrophoresis followed by autoradiog-raphy as described in Materials and Methods. Most of theearly proteins made in ¢149-infected V. cholerae biotypeclassical cells (15) were also made in the ¢149-infectedbiotype el tor cells. It is significant that the 64,500-daltonearly protein (E9), which is required for the synthesis of theconcatemeric intermediate, is made in (149-infected V.cholerae biotype el tor cells (15). Some of the late phage-specific proteins were also synthesized in infected V. chol-erae biotype el tor cells at 30 min after infection, when thephage-specific DNAs were resolved in gradients as concate-mers. However, the amount of these proteins was graduallyreduced as infection progressed, parallel to the degradationof the high-molecular-weight DNA.

DISCUSSION

The results presented in this report show that the nonper-missiveness of V. cholerae biotype el tor cells to (149infection is not due to a defect in adsorption or restriction ofthe injected phage DNA, since the parental DNA was stablein these cells until 60 min of infection.

Analysis of intracellular replication of (149 in V. choleraebiotype el tor cells has shown that the DNA synthesis atearly times after infection (up to about 20 min) takes place ina manner similar to that observed in classical vibrios (3), andmonomeric DNA molecules are synthesized (Fig. 3). At latertimes during infection, although some high-molecular-weightDNA intermediates were synthesized in V. cholerae biotypeel tor cells, these DNA molecules were unstable and couldnot be chased to monomeric units (Fig. 4). These resultssuggest two possibilities. First, the high-molecular-weightDNA intermediate may be defective and cannot serve as asubstrate for the synthesis of mature phage DNA. On theother hand, the high-molecular-weight DNA intermediatemay be a competent substrate but its maturation to monomersize, which presumably requires prohead formation andDNA packaging, fails to take place due to the deficiency inlate protein synthesis. On the basis of the results presentedin this report it is difficult to distinguish between these twopossibilities. These high-molecular-weight DNA intermedi-ates were resolved in the gradient as a broad peak (Fig. 3),compared with a much sharper peak obtained for infection ofclassical vibrios (3), indicating that molecules of varioussizes are produced. Furthermore, although 50% of the newlysynthesized concatemeric DNA molecules were associatedwith the cell membrane during infection of V. choleraebiotype classical cells by (149, not more than 10% of thehigh-molecular-weight DNA formed during infection of V.cholerae biotype el tor cells was membrane associated.Whether membrane association of the concatemeric DNA isnecessary for the stability of this DNA replicative interme-diate and for the subsequent cleavage into monomers andpackaging into phage heads is not known.

It has recently been reported based on the ultrastructuralanalysis of (149-infected classical and el tor cells that at

early times (5 min) after infection of V. cholerae biotype eltor strain cells, the cell DNA was completely degraded. Upto 60 min after infection there was no change in the ultra-structure of the infected cells, after which lysis of some ofthe cells occured without any release of infective particles(10). The concatemeric DNA replicative intermediates

VOL. 63, 1989 395

396 CHOWDHURY ET AL.

ab c d

~LE4|1 7II5E8^=

FIG. 5. Autoradiograph of [35S]methionine-labeled 44149 pro-

teins synthesized in UV-irradiated V. cholerae biotype el tor strain

MAK757 cells. Cells in the logarithmic phase of growth were UV

irradiated and infected with 44149 at an MOI of 10 as described in the

text. At 0 (a), 30 (b), 40 (c), and 50 (d) min after infection, samples

(1 ml) were removed and labeled with 10 p.Ci of [35S]methionine for

10 min. Labeling was terminated, and cells were lysed and analyzed

by sodium dodecyl sulfate-polyacrylamide gel electrophoresis fol-

lowed by autoradiography of dried gels as described in the text. E

and L represent early and late proteins, respectively, as described

previously (15).

formed during infection of V. cholerae biotype el tor cellsbeing degraded (Fig. 4) were not seen in thin sections ofinfected cells. Under identical experimental conditions con-catemeric DNA could be resolved in thin sections of infectedclassical V. cholerae cells (10). The results presented in thisreport are in agreement with the electron microscopic ob-servation.

It has been shown that for infection by 4149 the synthesisof the concatemeric DNA intermediate is coupled to thesynthesis of phage-specific late proteins. Under conditions inwhich concatemer synthesis is blocked, although most of theearly proteins are made, none of the late proteins is synthe-sized (3, 15). The results presented here show that forinfection of V. cholerae biotype el tor cells by phage 4149the late proteins appeared in infected cells when the high-molecular-weight DNA intermediates could be resolved.However, although in 4149-infected classical V. choleraecells, the synthesis of late proteins continued unperturbeduntil 60 min after infection, in the V. cholerae biotype el torcells after 30 min of infection, synthesis of late proteins wasgradually reduced concomitant with the degradation of theconcatemeric DNA intermediates. This result suggests thatthe presence of stable concatemers may be necessary for thesynthesis of late proteins. Interestingly enough, the 64,500-dalton early protein (E9), which is required for the synthesisof the concatemeric DNA intermediate, was made in 4149-infected V. cholerae biotype el tor cells (Fig. 5), suggestingthat this protein is required for the synthesis but not for thestabilization of the concatemeric DNA.The difference in the stability of concatemeric intermedi-

ates during intracellular replication of phage 4149 in V.cholerae biotype classical and el tor cells reflects some

inherent property of the host system. One such propertymay be the status of the recBC gene in the two biotypes. The

recBC status of host cells is known to influence productivephage infection in certain systems. It has been reported thatP1 and P2 phages fail to produce PFUs in recBC mutanthosts, although host recombination functions do not seem tobe required during the early phase of P1 replication (19).However, T4 gene 2 mutants do not grow in recBC+ cells butgrow in recBC mutants (13). From studies on DNA repairmechanisms operative in classical vibrios, it has been pre-dicted that although these cells have a functional recA gene(14) they might lack the recBC enzyme (6, 7). The status ofthe recBC system in V. cholerae biotype el tor has not beenexamined as yet. If for stabilization of 4149 concatemericDNA intermediates a protein has to bind to the concatemerto protect it from recBC-mediated degradation and such aprotein is not coded for by the 4149 genome, the concate-mers will be degraded in a recBC+ background. The heter-ogeneity in the sizes of the high-molecular-weight DNAintermediates and their rapid degradation with time in V.cholerae biotype el tor cells suggest such a possibility.

ACKNOWLEDGMENTS

We thank I. Guha Thakurta for excellent technical assistance. Weare grateful to all the members of the Biophysics Division for theirkind cooperation and encouragement during the study.The work was supported by the Council of Scientific and Indus-

trial Research and the Department of Science and Technology (grantSPISO/D-64/85), Government of India.

LITERATURE CITED

1. Chatterjee, S. N., J. Das, and D. Barua. 1965. Electron micros-copy of choleraphages. Indian J. Med. Res. 53:934-937.

2. Chowdhury, R., and J. Das. 1986. Infection by choleraphage0138: bacteriophage DNA and replicative intermediates. J.Virol. 57:960-967.

3. Chowdhury, R., A. Sengupta, P. Ray, and J. Das. 1987. Repli-cation and packaging of choleraphage 4149 DNA. J. Virol. 61:3999-4006.

4. Citarella, R. V., and R. R. Colwell. 1970. Polyphasic taxonomyof the genus Vibrio: polynucleotide sequence relationshipsamong selected Vibrio species. J. Bacteriol. 104:434 442.

5. Colwell, R. R. 1970. Polyphasic taxonomy of genus Vibrio:numerical taxonomy of Vibrio cholerae, Vibrio parahaemoly-ticus, and related Vibrio species. J. Bacteriol. 104:410-433.

6. Das, G., and J. Das. 1983. Radiation sensitive mutant ofhypertoxinogenic strain 569B of Vibrio cholerae. Mutat. Res.109:21-30.

7. Das, G., K. Sil, and J. Das. 1981. Repair of ultraviolet lightinduced DNA damage in Vibrio cholerae. Biochim. Biophys.Acta 655:413-420.

8. Finkelstein, R. A. 1973. Cholera. Crit. Rev. Microbiol. 3:552-623.

9. Lohia, A., S. Majumdar, A. N. Chatterjee, and J. Das. 1985.Effect of changes in the osmolarity of the growth medium onVibrio cholerae cells. J. Bacteriol. 163:1158-1166.

10. Majumdar, S., S. N. Dey, R. Chowdhury, C. Dutta, and J. Das.1988. Intracellular development of choleraphage 4149 underpermissive and non-permissive conditions: an electron micro-scopic study. Intervirology 29:27-38.

11. Mukherjee, S. 1963. Bacteriophage typing of cholera. Bull.W.H.O. 28:337-345.

12. Mukherjee, S. 1978. Principles and practice of typing Vibriocholerae. Methods Microbiol. 12:74-115.

13. Oliver, D. B., and E. B. Goldberg. 1977. Protection of parentalT4 DNA from a restriction exonuclease by the product of gene2. J. Mol. Biol. 116:877-881.

14. Paul, K., S. K. Ghosh, and J. Das. 1986. Cloning and expressionin Escherichia coli of a recA like gene from Vibrio cholerae.Mol. Gen. Genet. 203:58-61.

J. VIROL.

ABORTIVE 0149 REPLICATION IN V. CHOLERAE BIOTYPE EL TOR

15. Ray, P., A. Sengupta, and J. Das. 1984. Phosphate repression ofphage protein synthesis during infection by choleraphage 4149.Virology 136:110-124.

16. Roy, N. K., G. Das, T. S. Balganesh, S. N. Dey, R. K. Ghosh, andJ. Das. 1982. Enterotoxin production, DNA repair and alkalinephosphate of Vibrio cholerae before and after animal passage. J.Gen. Microbiol. 128:1927-1932.

17. Sengupta, A., P. Ray, and J. Das. 1985. Characterisation andphysical map of the choleraphage 4)149 DNA. Virology 140:

217-229.18. Takeya, K., Y. Zinnaka, S. Shimodori, Y. Nakayama, K.

Amako, and I. Kyoko. 1965. Lysogeny in El Tor vibrio, p. 24-29. In Cholera Research Symposium proceedings. U.S. Depart-ment of Health, Education and Welfare, Public Health Service,Washington, D.C.

19. Zabrovitz, S., N. Sergev, and G. Cohen. 1977. Growth ofbacteriophage P1 in recombination deficient hosts of E. coli.Virology 80:233-248.

VOL. 63, 1989 397